CN106992007B - Data processing method and device based on voice recognition scoring system - Google Patents

Data processing method and device based on voice recognition scoring system Download PDF

Info

Publication number
CN106992007B
CN106992007B CN201710193915.2A CN201710193915A CN106992007B CN 106992007 B CN106992007 B CN 106992007B CN 201710193915 A CN201710193915 A CN 201710193915A CN 106992007 B CN106992007 B CN 106992007B
Authority
CN
China
Prior art keywords
frame
data
processing
thread
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201710193915.2A
Other languages
Chinese (zh)
Other versions
CN106992007A (en
Inventor
高永虎
彭一平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Baidu Netcom Science and Technology Co Ltd
Original Assignee
Beijing Baidu Netcom Science and Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Baidu Netcom Science and Technology Co Ltd filed Critical Beijing Baidu Netcom Science and Technology Co Ltd
Priority to CN201710193915.2A priority Critical patent/CN106992007B/en
Publication of CN106992007A publication Critical patent/CN106992007A/en
Priority to US15/729,680 priority patent/US10573306B2/en
Application granted granted Critical
Publication of CN106992007B publication Critical patent/CN106992007B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/28—Constructional details of speech recognition systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Telephonic Communication Services (AREA)
  • Document Processing Apparatus (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention provides a data processing method and a data processing device based on a voice recognition scoring system, wherein the data processing method based on the voice recognition scoring system comprises the steps of obtaining data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; performing frame splicing processing on multi-frame data, and taking a data frame subjected to frame splicing processing as a target data frame; a target thread is determined from the plurality of threads and a target data frame is processed based on the target thread. The invention can improve the data processing efficiency of the voice recognition-based scoring system and save system resources consumed by processing logic.

Description

Data processing method and device based on voice recognition scoring system
Technical Field
The invention relates to the technical field of voice recognition, in particular to a data processing method and device based on a voice recognition scoring system.
Background
According to the working principle of the voice recognition scoring system in the related art, after voices of multiple users are received, voice feature data can be extracted aiming at the voice of each user, then multiple threads are started, each thread correspondingly processes the voice feature data of each user, and an acoustic model score is generated for the voice feature data so as to apply the acoustic model score to a subsequent voice recognition system. A scheduling module is preset in the voice recognition scoring system, and schedules a plurality of threads based on computing resources of a server/electronic equipment so that the plurality of threads serially process corresponding voice characteristic data.
In this way, each thread needs to execute a complete data processing flow, and needs to uniformly schedule multiple threads, which causes a large delay in data processing, high resource consumption, and low data processing efficiency.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, an object of the present invention is to provide a data processing method based on a speech recognition scoring system, which can improve the data processing efficiency based on the speech recognition scoring system and save system resources consumed by processing logic.
Another object of the present invention is to provide a data processing apparatus based on a voice recognition scoring system.
It is a further object of the present invention to provide a data processing apparatus based on a speech recognition scoring system.
It is another object of the invention to propose a non-transitory computer-readable storage medium.
It is a further object of the invention to propose a computer program product.
In order to achieve the above object, a data processing method based on a speech recognition scoring system according to an embodiment of a first aspect of the present invention includes: acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame; determining a target thread from the plurality of threads and processing the target data frame based on the target thread.
In the data processing method based on the speech recognition scoring system provided in the embodiment of the first aspect of the present invention, the data frame to be processed in each thread is subjected to frame splicing processing, the data frame subjected to frame splicing processing is used as a target data frame, a target thread is determined from a plurality of threads, the target data frame is processed based on the target thread, and by performing frame splicing processing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling of the plurality of threads is not required, so that the data processing efficiency based on the speech recognition scoring system can be improved, and system resources consumed by processing logic are saved.
In order to achieve the above object, a data processing apparatus based on a speech recognition scoring system according to a second embodiment of the present invention includes: the first acquisition module is used for acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; the frame splicing processing module is used for carrying out frame splicing processing on the multi-frame data and taking the data frame subjected to frame splicing processing as a target data frame; and the processing module is used for determining a target thread from the multiple threads and processing the target data frame based on the target thread.
According to the data processing device based on the voice recognition scoring system provided by the embodiment of the second aspect of the invention, the data frame to be processed of each thread is subjected to frame splicing processing, the data frame subjected to frame splicing processing is taken as a target data frame, a target thread is determined from a plurality of threads, the target data frame is processed based on the target thread, and by the frame splicing processing, the situation that each thread needs to execute a complete data processing flow is avoided, and unified scheduling of the plurality of threads is not needed, so that the data processing efficiency based on the voice recognition scoring system can be improved, and system resources consumed by processing logic are saved.
In order to achieve the above object, a data processing apparatus based on a speech recognition scoring system according to a third aspect of the present invention includes: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to: acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame; determining a target thread from the plurality of threads and processing the target data frame based on the target thread.
In the data processing apparatus based on the speech recognition scoring system according to the third aspect of the present invention, the data frame to be processed in each thread is subjected to frame splicing processing, the data frame subjected to frame splicing processing is used as a target data frame, a target thread is determined from a plurality of threads, the target data frame is processed based on the target thread, and by performing frame splicing processing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling is not required for the plurality of threads.
To achieve the above object, a non-transitory computer-readable storage medium according to a fourth aspect of the present invention is a non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of a mobile terminal, the instructions enabling the mobile terminal to execute a data processing method based on a voice recognition scoring system, the method including: acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame; determining a target thread from the plurality of threads and processing the target data frame based on the target thread.
The non-transitory computer-readable storage medium according to the fourth aspect of the present invention determines a target thread from a plurality of threads by performing frame splicing processing on a data frame to be processed in each thread, and using the data frame subjected to frame splicing processing as a target data frame, and processes the target data frame based on the target thread, and by performing frame splicing processing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling is not required for the plurality of threads, so that data processing efficiency of a scoring system based on speech recognition can be improved, and system resources consumed by processing logic are saved.
To achieve the above object, a computer program product according to a fifth embodiment of the present invention is a computer program product, which when executed by an instruction processor performs a data processing method based on a speech recognition scoring system, the method including: acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data; performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame; determining a target thread from the plurality of threads and processing the target data frame based on the target thread.
In the computer program product provided in the embodiment of the fifth aspect of the present invention, by performing frame splicing on a data frame to be processed in each thread, and taking the data frame subjected to frame splicing as a target data frame, a target thread is determined from a plurality of threads, and the target data frame is processed based on the target thread, and by performing frame splicing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling is not required for a plurality of threads, so that data processing efficiency of a scoring system based on speech recognition can be improved, and system resources consumed by processing logic are saved.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a schematic flow chart of a data processing method based on a speech recognition scoring system according to an embodiment of the present invention;
FIG. 2 is a flow chart of a data processing method based on a voice recognition scoring system according to another embodiment of the present invention;
FIG. 3 is a schematic diagram of a data processing method scenario based on a speech recognition scoring system according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a data processing apparatus based on a voice recognition scoring system according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a data processing apparatus based on a speech recognition scoring system according to another embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention. On the contrary, the embodiments of the invention include all changes, modifications and equivalents coming within the spirit and terms of the claims appended hereto.
Fig. 1 is a flowchart illustrating a data processing method based on a speech recognition scoring system according to an embodiment of the present invention.
The data processing method based on the voice recognition scoring system in the present embodiment may be configured in a data processing apparatus based on the voice recognition scoring system.
The data processing apparatus based on the voice recognition scoring system may be disposed in a server, or may also be disposed in an electronic device, which is not limited in this embodiment of the present invention.
The electronic device is, for example, a Personal Computer (PC), a cloud device or a mobile device, and the mobile device is, for example, a smart phone or a tablet Computer.
It should be noted that the execution main body of the embodiment of the present invention may be, for example, a Central Processing Unit (CPU) in the server/electronic device in terms of hardware, and may be, for example, an application program in the server/electronic device in terms of software, which is not limited to this.
The data processing method based on the voice recognition scoring system can be applied to the process of generating the acoustic model score for the voice characteristic data by taking the voice characteristic data as input based on the acoustic model in the voice recognition.
Referring to fig. 1, the method includes:
s11: and acquiring data frames to be processed of each thread in the multiple threads to obtain multi-frame data.
In the embodiment of the invention, based on the working principle of the voice recognition scoring system, after a plurality of threads are started and voice characteristic data frames of corresponding users are distributed to each thread, data frames to be processed of each thread in the plurality of threads can be obtained, and multi-frame data is obtained.
S12: and performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame.
Wherein, piece together frame processing to multiframe data includes:
and performing splicing frame processing in a cross arrangement mode on the multi-frame data.
Through carrying out the frame splicing processing in a cross arrangement mode on multi-frame data, the parallel processing of the thread on the data in the target data frame can be guaranteed, and the processing efficiency is improved on the premise of not influencing the processing accuracy.
In the related art, each thread correspondingly processes a voice feature data frame of each user, and generates an acoustic model score for the voice feature data, so as to apply the acoustic model score in a subsequent voice recognition system.
In the embodiment of the invention, after obtaining the data frames to be processed of each thread in the multiple threads, the multi-frame data is subjected to frame splicing processing, and by performing the frame splicing processing, the situation that each thread needs to execute a complete data processing flow is avoided, and the multiple threads do not need to be uniformly scheduled, so that the data processing efficiency of the voice recognition scoring system can be improved, and the system resources consumed by processing logic are saved.
Optionally, the number of the target data frames is equal to the product of the number of the data frames to be processed by each thread and the number of the threads.
For example, if the number of data frames to be processed in each thread is 4 and there are 3 threads, the multi-frame data is subjected to frame splicing, and the number of target data frames subjected to frame splicing is 12.
S13: a target thread is determined from the plurality of threads and a target data frame is processed based on the target thread.
In the embodiment of the present invention, the target thread may be determined based on the load capacity of the thread, that is, a thread capable of carrying a processing target data frame among the plurality of threads is taken as the target thread, or the target thread may be determined randomly, that is, any one of the plurality of threads is taken as the target thread, which is not limited in this respect.
After the target thread is determined, other threads can be mounted.
In the embodiment of the invention, the computing resources of the server/electronic device can be called based on the target thread to execute the logic operation for processing the target data frame, and because the target thread processes the target data frame, the situation that each thread needs to execute a complete data processing flow is avoided.
In this embodiment, by performing frame splicing processing on the data frame to be processed by each thread, and taking the data frame subjected to frame splicing processing as a target data frame, a target thread is determined from a plurality of threads, and the target data frame is processed based on the target thread, and by performing frame splicing processing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling of a plurality of threads is not required, so that the data processing efficiency of a scoring system based on speech recognition can be improved, and system resources consumed by processing logic are saved.
Fig. 2 is a flowchart illustrating a data processing method based on a speech recognition scoring system according to another embodiment of the present invention.
Referring to fig. 2, the method includes:
s21: and acquiring data frames to be processed of each thread in the multiple threads to obtain multi-frame data.
In the embodiment of the invention, based on the working principle of the voice recognition scoring system, after a plurality of threads are started and voice characteristic data frames of corresponding users are distributed to each thread, data frames to be processed of each thread in the plurality of threads can be obtained, and multi-frame data is obtained.
In this embodiment, the number of threads is 3, and the number of frames of the data frame to be processed corresponding to each thread is 4.
Referring to fig. 3, fig. 3 is a schematic view of a data processing method based on a speech recognition scoring system in an embodiment of the present invention, where the data processing method includes: 3 threads 31, a target data frame 32, and a processing result 33.
S22: and allocating corresponding zone bits to the data frames to be processed of each thread.
In an embodiment of the present invention, the flag bit is used to uniquely identify the data frame to be processed by each thread, and the flag bit may be, for example, an identity authentication Identifier (ID) corresponding to the thread.
By allocating corresponding zone bits to the data frames to be processed of each thread in advance, the corresponding processing results can be read from the processing results based on the zone bits by the threads except the target thread in the multiple threads, and the consistency of the acoustic model scores in the subsequent speech recognition system application is guaranteed.
For example, referring to FIG. 3, the ID of a thread may be 0, 1, 2, respectively. The 4 frame data frames to be processed in each thread may be acquired, and the ID of the corresponding thread is respectively allocated to the data frames to be processed in each thread, that is, the flag bit of the 4 frame data frame to be processed in thread No. 0 is 0, the flag bit of the 4 frame data frame to be processed in thread No. 1 is 1, and the flag bit of the 4 frame data frame to be processed in thread No. 2 is 2.
S23: and performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame.
Wherein, piece together frame processing to multiframe data includes:
and performing splicing frame processing in a cross arrangement mode on the multi-frame data.
Through carrying out the frame splicing processing in a cross arrangement mode on multi-frame data, the parallel processing of the thread on the data in the target data frame can be guaranteed, and the processing efficiency is improved on the premise of not influencing the processing accuracy.
In the related art, each thread correspondingly processes a voice feature data frame of each user, and generates an acoustic model score for the voice feature data, so as to apply the acoustic model score in a subsequent voice recognition system.
In the embodiment of the invention, after obtaining the data frames to be processed of each thread in the multiple threads, the multi-frame data is subjected to frame splicing processing, and by performing the frame splicing processing, the situation that each thread needs to execute a complete data processing flow is avoided, and the multiple threads do not need to be uniformly scheduled, so that the data processing efficiency of the voice recognition scoring system can be improved, and the system resources consumed by processing logic are saved.
Optionally, the number of the target data frames is equal to the product of the number of the data frames to be processed by each thread and the number of the threads.
For example, if the number of data frames to be processed in each thread is 4 and there are 3 threads, the multi-frame data is subjected to frame splicing, and the number of target data frames subjected to frame splicing is 12.
Referring to the target data frame 32 in fig. 3, the multi-frame data may be subjected to a frame splicing process, which may be, for example, a cross arrangement, for example, a first frame data of 4 frame data frames with a flag bit of 0 is used as a first frame of the target data frame, a first frame data of 4 frame data frames with a flag bit of 1 is used as a second frame of the target data frame, a first frame data of 4 frame data frames with a flag bit of 2 is used as a third frame of the target data frame, a second frame data of 4 frame data frames with a flag bit of 0 is used as a fourth frame of the target data frame, a second frame data of 4 frame data frames with a flag bit of 1 is used as a fifth frame of the target data frame, a second frame data of 4 frame data frames with a flag bit of 2 is used as a sixth frame of the target data frame, and so on.
S24: and marking the data frame to be processed of each thread corresponding to the flag bit in the target data frame.
For example, referring to fig. 3, the first frame data, the fourth frame data, the seventh frame data, and the tenth frame data in the target data frame may be labeled with a flag bit 0, while the second frame data, the fifth frame data, the eighth frame data, and the eleventh frame data in the target data frame may be labeled with a flag bit 1, and the third frame data, the sixth frame data, the ninth frame data, and the twelfth frame data in the target data frame may be labeled with a flag bit 2.
S25: a target thread is determined from the plurality of threads and a target data frame is processed based on the target thread.
In the embodiment of the present invention, the target thread may be determined based on the load capacity of the thread, that is, a thread capable of carrying a processing target data frame among the plurality of threads is taken as the target thread, or the target thread may be determined randomly, that is, any one of the plurality of threads is taken as the target thread, which is not limited in this respect.
After the target thread is determined, other threads can be mounted.
In the embodiment of the invention, the computing resources of the server/electronic device can be called based on the target thread to execute the logic operation for processing the target data frame, and because the target thread processes the target data frame, the situation that each thread needs to execute a complete data processing flow is avoided.
Referring to fig. 3, the target thread may be, for example, thread number 0.
S26: and acquiring a processing result of processing the target data frame.
In the embodiment of the present invention, a processing result of the target thread processing the target data frame may be obtained.
S27: and reading the corresponding processing result from the processing result by the threads except the target thread in the plurality of threads based on the flag bit.
S28: and taking the corresponding processing result as the result of processing the data frame to be processed by each thread.
For example, referring to processing result 33 in fig. 3, thread No. 1 may read the processing result corresponding to flag bit 1 from the processing result of the target thread, thread No. 2 may read the processing result corresponding to flag bit 2 from the processing result of the target thread, and the remaining result in the processing result of thread No. 0 is the processing result corresponding to flag bit 0.
The corresponding processing result is read from the processing result based on the flag bit by the threads except the target thread in the multiple threads, and the corresponding processing result is used as the result of processing the data frame to be processed by each thread, so that the data processing structure design based on the voice recognition scoring system is simplified, the multiple threads are not required to be uniformly scheduled, the data processing logic is executed by a single thread, and other threads directly read the corresponding processing result based on the flag bit, and the computing resources of the processor are effectively saved.
In this embodiment, by allocating the corresponding flag bit to the data frame to be processed by each thread in advance, the corresponding processing result can be read from the processing result based on the flag bit for the threads other than the target thread among the multiple threads, so that the continuity of the acoustic model score in the application of the subsequent speech recognition system is ensured. Through carrying out the frame splicing processing in a cross arrangement mode on multi-frame data, the parallel processing of the thread on the data in the target data frame can be guaranteed, and the processing efficiency is improved on the premise of not influencing the processing accuracy. By performing frame splicing processing on the data frame to be processed of each thread and taking the data frame subjected to frame splicing processing as a target data frame, a target thread is determined from a plurality of threads, the target data frame is processed based on the target thread, and by performing frame splicing processing, the situation that each thread needs to execute a complete data processing flow is avoided, and unified scheduling on the plurality of threads is not needed, so that the data processing efficiency of a scoring system based on voice recognition can be improved, and system resources consumed by processing logic are saved. The corresponding processing result is read from the processing result based on the flag bit by the threads except the target thread in the multiple threads, and the corresponding processing result is used as the result of processing the data frame to be processed by each thread, so that the data processing structure design based on the voice recognition scoring system is simplified, the multiple threads are not required to be uniformly scheduled, the data processing logic is executed by a single thread, and other threads directly read the corresponding processing result based on the flag bit, and the computing resources of the processor are effectively saved.
Fig. 4 is a schematic structural diagram of a data processing apparatus based on a speech recognition scoring system according to an embodiment of the present invention.
Referring to fig. 4, the data processing apparatus 400 based on the voice recognition scoring system includes: a first acquisition module 401, a frame-splicing processing module 402, and a processing module 403, wherein,
the first obtaining module 401 is configured to obtain a data frame to be processed by each thread in a plurality of threads, so as to obtain multi-frame data.
The frame splicing processing module 402 is configured to perform frame splicing processing on multiple frames of data, and use the data frame after frame splicing processing as a target data frame.
Optionally, the number of the target data frames is equal to the product of the number of the data frames to be processed by each thread and the number of the threads.
Optionally, the frame splicing processing module 402 is specifically configured to:
and performing splicing frame processing in a cross arrangement mode on the multi-frame data.
A processing module 403, configured to determine a target thread from the multiple threads, and process the target data frame based on the target thread.
Optionally, in some embodiments, referring to fig. 5, the data processing apparatus 400 based on the speech recognition scoring system further includes:
an allocating module 404, configured to allocate a corresponding flag bit to the data frame to be processed by each thread.
A marking module 405, configured to mark a data frame to be processed by each thread corresponding to the flag bit in the target data frame.
A second obtaining module 406, configured to obtain a processing result of processing the target data frame.
The reading module 407 is configured to enable threads other than the target thread among the multiple threads to read a corresponding processing result from the processing result based on the flag bit, and use the corresponding processing result as a result of processing the data frame to be processed by each thread.
It should be noted that the explanation of the embodiment of the data processing method based on the speech recognition scoring system in the foregoing fig. 1-fig. 3 is also applicable to the data processing apparatus 400 based on the speech recognition scoring system in the embodiment, and the implementation principle thereof is similar and will not be described herein again.
In this embodiment, by performing frame splicing processing on the data frame to be processed by each thread, and taking the data frame subjected to frame splicing processing as a target data frame, a target thread is determined from a plurality of threads, and the target data frame is processed based on the target thread, and by performing frame splicing processing, it is avoided that each thread needs to execute a complete data processing flow, and uniform scheduling of a plurality of threads is not required, so that the data processing efficiency of a scoring system based on speech recognition can be improved, and system resources consumed by processing logic are saved.
It should be noted that the terms "first," "second," and the like in the description of the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In addition, in the description of the present invention, "a plurality" means two or more unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.

Claims (10)

1. A data processing method based on a voice recognition scoring system is characterized by comprising the following steps:
acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data;
performing frame splicing processing on the multi-frame data, and taking the data frame subjected to frame splicing processing as a target data frame;
determining a target thread from the plurality of threads and processing the target data frame based on the target thread;
in the acquiring of the data frame to be processed by each thread in the plurality of threads, after obtaining the multi-frame data, the method includes:
and distributing corresponding zone bits to the data frame to be processed of each thread.
2. The data processing method based on the speech recognition scoring system as claimed in claim 1, wherein after the frame splicing processing is performed on the multi-frame data and the frame-spliced data frame is taken as a target data frame, the method comprises:
and marking the data frame to be processed of each thread corresponding to the flag bit in the target data frame.
3. The data processing method based on the speech recognition scoring system of claim 2, wherein after the determining a target thread from the plurality of threads and processing the target data frame based on the target thread, further comprising:
acquiring a processing result of processing the target data frame;
reading corresponding processing results from the processing results by the threads except the target thread in the plurality of threads based on the flag bits;
and taking the corresponding processing result as a result of processing the data frame to be processed by each thread.
4. The data processing method based on the speech recognition scoring system as recited in claim 1, wherein the frame splicing processing of the multi-frame data comprises:
and performing splicing frame processing in a cross arrangement mode on the multi-frame data.
5. The data processing method based on the speech recognition scoring system according to any one of claims 1-4, wherein the number of the target data frames is equal to the product of the number of the data frames to be processed by each thread and the number of the threads.
6. A data processing apparatus based on a speech recognition scoring system, comprising:
the first acquisition module is used for acquiring data frames to be processed of each thread in a plurality of threads to obtain multi-frame data;
the frame splicing processing module is used for carrying out frame splicing processing on the multi-frame data and taking the data frame subjected to frame splicing processing as a target data frame;
the processing module is used for determining a target thread from the multiple threads and processing the target data frame based on the target thread;
the method comprises the following steps:
and the distribution module is used for distributing corresponding zone bits to the data frames to be processed of each thread.
7. The data processing apparatus of claim 6, comprising:
and the marking module is used for marking the data frame to be processed of each thread corresponding to the flag bit in the target data frame.
8. The data processing apparatus based on a speech recognition scoring system of claim 7, further comprising:
the second acquisition module is used for acquiring a processing result of processing the target data frame;
a reading module, configured to enable threads other than the target thread in the multiple threads to read a corresponding processing result from the processing result based on the flag bit, and use the corresponding processing result as a result of processing the to-be-processed data frame by each thread.
9. The speech recognition scoring system-based data processing apparatus of claim 6, wherein the framing processing module is specifically configured to:
and performing splicing frame processing in a cross arrangement mode on the multi-frame data.
10. The data processing apparatus of claim 6-9, wherein the number of target data frames is equal to the product of the number of data frames to be processed per thread and the number of threads.
CN201710193915.2A 2017-03-28 2017-03-28 Data processing method and device based on voice recognition scoring system Active CN106992007B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201710193915.2A CN106992007B (en) 2017-03-28 2017-03-28 Data processing method and device based on voice recognition scoring system
US15/729,680 US10573306B2 (en) 2017-03-28 2017-10-11 Method and device for processing data based on speech recognizing and scoring system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710193915.2A CN106992007B (en) 2017-03-28 2017-03-28 Data processing method and device based on voice recognition scoring system

Publications (2)

Publication Number Publication Date
CN106992007A CN106992007A (en) 2017-07-28
CN106992007B true CN106992007B (en) 2020-07-28

Family

ID=59413038

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710193915.2A Active CN106992007B (en) 2017-03-28 2017-03-28 Data processing method and device based on voice recognition scoring system

Country Status (2)

Country Link
US (1) US10573306B2 (en)
CN (1) CN106992007B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110136697B (en) * 2019-06-06 2021-03-30 深圳市数字星河科技有限公司 English reading practice system based on multi-process/thread parallel operation

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1503220A (en) * 2002-11-20 2004-06-09 中国科学院声学研究所 Parallel searching methd for speech recognition
CN1540508A (en) * 2003-04-24 2004-10-27 �Ҵ���˾ Dynamic switching of multiline range processor between single line range and complicated multiline range mode
CN101826263A (en) * 2009-03-04 2010-09-08 中国科学院自动化研究所 Objective standard based automatic oral evaluation system
CN101923854A (en) * 2010-08-31 2010-12-22 中国科学院计算技术研究所 Interactive speech recognition system and method
CN103915092A (en) * 2014-04-01 2014-07-09 百度在线网络技术（北京）有限公司 Voice identification method and device
CN104778074A (en) * 2014-01-14 2015-07-15 腾讯科技（深圳）有限公司 Calculation task processing method and device
CN106328135A (en) * 2016-08-19 2017-01-11 北京智能管家科技有限公司 Application switching method and application switching device based on voice

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7814486B2 (en) * 2006-06-20 2010-10-12 Google Inc. Multi-thread runtime system
US8261270B2 (en) * 2006-06-20 2012-09-04 Google Inc. Systems and methods for generating reference results using a parallel-processing computer system
US8146066B2 (en) * 2006-06-20 2012-03-27 Google Inc. Systems and methods for caching compute kernels for an application running on a parallel-processing computer system
US8650029B2 (en) * 2011-02-25 2014-02-11 Microsoft Corporation Leveraging speech recognizer feedback for voice activity detection
CN102223452B (en) * 2011-06-16 2016-12-07 中兴通讯股份有限公司 The TTY method of testing of mobile terminal and system
US9514747B1 (en) * 2013-08-28 2016-12-06 Amazon Technologies, Inc. Reducing speech recognition latency
GB2528265B (en) * 2014-07-15 2021-03-10 Advanced Risc Mach Ltd Method of and apparatus for generating an output frame
US20160093297A1 (en) * 2014-09-26 2016-03-31 Michael E. Deisher Method and apparatus for efficient, low power finite state transducer decoding
US10109277B2 (en) * 2015-04-27 2018-10-23 Nuance Communications, Inc. Methods and apparatus for speech recognition using visual information
CN105702257A (en) * 2015-08-12 2016-06-22 乐视致新电子科技（天津）有限公司 Speech processing method and device

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1503220A (en) * 2002-11-20 2004-06-09 中国科学院声学研究所 Parallel searching methd for speech recognition
CN1540508A (en) * 2003-04-24 2004-10-27 �Ҵ���˾ Dynamic switching of multiline range processor between single line range and complicated multiline range mode
CN101826263A (en) * 2009-03-04 2010-09-08 中国科学院自动化研究所 Objective standard based automatic oral evaluation system
CN101923854A (en) * 2010-08-31 2010-12-22 中国科学院计算技术研究所 Interactive speech recognition system and method
CN104778074A (en) * 2014-01-14 2015-07-15 腾讯科技（深圳）有限公司 Calculation task processing method and device
CN103915092A (en) * 2014-04-01 2014-07-09 百度在线网络技术（北京）有限公司 Voice identification method and device
CN106328135A (en) * 2016-08-19 2017-01-11 北京智能管家科技有限公司 Application switching method and application switching device based on voice

Also Published As

Publication number Publication date
CN106992007A (en) 2017-07-28
US10573306B2 (en) 2020-02-25
US20180286393A1 (en) 2018-10-04

Similar Documents

Publication Publication Date Title
US11146502B2 (en) Method and apparatus for allocating resource
US9871741B2 (en) Resource management based on device-specific or user-specific resource usage profiles
US9697124B2 (en) Systems and methods for providing dynamic cache extension in a multi-cluster heterogeneous processor architecture
US20160210174A1 (en) Hybrid Scheduler and Power Manager
JP2017514209A (en) Dynamic resource management for multi-process applications
CN112988362B (en) Task processing method and device, electronic equipment and storage medium
CN112380020A (en) Computing power resource allocation method, device, equipment and storage medium
CN110162393B (en) Task scheduling method, device and storage medium
US9519490B2 (en) Adaptive data synchronization
US11302303B2 (en) Method and device for training an acoustic model
US20170201566A1 (en) File downloading method, apparatus, and terminal device
CN105843819B (en) Data export method and device
US9075789B2 (en) Methods and apparatus for interleaving priorities of a plurality of virtual processors
US20170177255A1 (en) Method and electronic device for allocating dynamic memory resources
KR20140145748A (en) Method for allocating process in multi core environment and apparatus therefor
US10248321B1 (en) Simulating multiple lower importance levels by actively feeding processes to a low-memory manager
CN111338787B (en) Data processing method and device, storage medium and electronic device
CN111158867A (en) Time synchronization processing method, thread scheduling device and electronic equipment
CN106992007B (en) Data processing method and device based on voice recognition scoring system
US20160335127A1 (en) System and method for dynamic granularity control of parallelized work in a portable computing device (pcd)
US10028285B2 (en) Method, device and terminal for allocating network services to data channels
CN112860401A (en) Task scheduling method and device, electronic equipment and storage medium
CN116881003A (en) Resource allocation method, device, service equipment and storage medium
CN114579187B (en) Instruction distribution method and device, electronic equipment and readable storage medium
CN115754413A (en) Oscilloscope and data processing method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant